Use of Systematic Evidence Mapping to Rapidly Update a Human Health Assessment of Formaldehyde
On this page:
Background and Purpose: Human health assessments address the first two steps of a risk assessment, hazard identification and dose-response analysis, and involve review of the available evidence (i.e., studies of exposed humans, experimental studies in animals, and experiments informing the mechanisms by which the chemical may cause health effects) by subject matter experts. Increasingly, systematic review approaches are being applied to increase the rigor of these reviews. In late 2017, at the request of senior EPA leadership, work was suspended on a completed draft human health assessment on the health effects of inhaling formaldehyde, a high production volume chemical with many commercial and industrial uses. In early 2021, work on the formaldehyde assessment was unsuspended and there was an urgent need to update the stalled assessment to meet the timing needs of EPA program offices. A systematic evidence map (SEM) was developed to rapidly update the literature base. SEMs apply the principles of systematic review to compile and present the available evidence on a topic, without synthesizing or drawing conclusions about that evidence.
Methods: To update the formaldehyde-specific literature, a series of searches were conducted for the time-period of 2016-2021, intentionally overlapping the dates searched to develop the 2017 draft. Two independent screeners applied Populations, Exposures, Comparators, Outcomes (PECO) criteria to screen the literature using DistillerSR software. Literature trees were developed in the Health Assessment Workspace Collaborative (HAWC) to summarize the screening decisions and studies were inventoried (i.e., organized by study features such as health effect or mechanism, species, etc.). Considerations were then applied to the studies meeting the PECO criteria in order to identify studies that could materially impact the judgments drawn in the 2017 draft assessment, specifically the identified hazards or the derived toxicity values. Examples of such considerations included preference for longer formaldehyde exposure durations and formaldehyde sources without significant concern for confounding. The studies identified as possibly impactful to the 2017 draft conclusions were incorporated into the updated draft evidence syntheses, while the SEM served as the record of decisions on the other identified studies.
Results: The SEM identified more than 13,000 studies over the 2016-2021 timeframe. Nearly 180 of these studies met the PECO criteria. Fifty-five of the studies meeting the PECO criteria were deemed possibly impactful and were incorporated into the evidence syntheses presented in the updated draft assessment. In general, these newer studies strengthened the hazard identification decisions presented in the 2017 draft assessment, although there were exceptions. The applied methods and decisions were compiled into an SEM Appendix to allow for easy and transparent referencing of the new literature considered in updating the draft assessment.
Conclusion: In mid-2021, a little over six months after work on the assessment was unsuspended, the updated draft began a multiple-step review process, which is ongoing. Use of an SEM approach allowed for the systematic identification and incorporation of potentially impactful literature into the assessment in a short timeframe to facilitate timely decision-making. This exemplifies that SEM approaches can provide a rapid, but still rigorous and transparent, means of updating complex assessments that inform decisions to protect human health.
Disclaimer: The views presented are those of the authors and do not necessarily represent the views or policies of the U.S. EPA.